Resting state MRI (rsMRI), based on fluctuations in the blood oxygenation level dependent (BOLD) signal, is increasingly used to map networks of spontaneous activity in the brain. The neural basis of these fluctuations is not well understood, with various studies reporting a link to low frequency power, high frequency power, modulation of spiking, and vasomotion. While the frequency range of the BOLD fluctuations is 0-0.1 Hz, previous studies have examined electrical activity in higher frequency bands (>1 Hz). It is known, however, that infra-slow oscillations (IFSOs; <1 Hz) exist in the brain and they have been linked to fluctuations in attentional control and reaction time in normal subjects and ADHD patients. We hypothesize that the BOLD fluctuations have a direct link to electrical fluctuations in the same frequency band, and that the modulation of higher frequencies by these slower oscillations leads to state-dependent relationships with the BOLD signal. 1. Determine the relationship between infra-slow potential fluctuations and activity in typical LFP bands (1-100 Hz). IFSOs and broadband local field potentials (LFPs) will be recorded from a network of cortical and subcortical sites to determine the spatial distribution of IFSOs and how they affect local activity. Simultaneous IFSO and intracellular recording will examine whether membrane potential changes are tied to low frequency oscillations. Different anesthetic states will modulate neural activity. 2. Characterize the contribution of IFSOs to the BOLD signal on a site-by-site and network basis. No studies have looked at the direct frequency correlates of the low frequency BOLD fluctuations. Preliminary data suggests that patterns of IFSOs can be mapped using MRI. Using a simultaneous recording/imaging protocol developed in our lab, we will obtain LFPs (broadband and infra-slow) and BOLD from sites selected from the network examined in aim 1. Correlation between LFPs and local BOLD signal will be performed to determine the largest contribution to BOLD fluctuations, while coherence between band-limited LFPs and BOLD correlation will be compared to identify the best predictors of BOLD correlation. 3. Examine the spatiotemporal dynamics of IFSOs and determine their link to quasi-periodic BOLD fluctuations. Preliminary data indicates that the time-lagged correlation between BOLD and IFSOs demonstrates a pattern of propagation along the cortex that is highly similar to the spatiotemporal dynamics previously observed with the BOLD signal. This aim will directly examine the link between BOLD and IFSO dynamics using the simultaneously-acquired multi-site data collected for aims 1 and 2. This project will provide unique insight into the network activity that underlies functional connectivity maps created with MRI and, if our hypothesis proves correct, will lead to a new way to map the spatiotemporal patterns of the infra-slow activity that modulates attention throughout the whole brain with resolution unobtainable with electroencephalography. PUBLIC HEALTH RELEVANCE: The goal of this project is to determine how very low frequency electrical activity modulates other activity in the brain, and how this in turn relates t the functional MRI signal. An improved understanding of spontaneous MRI signal fluctuations is the first step toward our long-term goal of establishing resting state MRI as a noninvasive tool for the diagnosis and evaluation of clinical disorders.